Modular adaptive design for active suspensions

Abstract

Active suspension designs rely on knowledge of system parameters such as vehicle mass to achieve their objectives of improved passenger comfort and road handling. Previously, we proposed a novel nonlinear design methodology for active suspensions, which uses a nonlinear control objective to explicitly incorporate rattlespace travel limits and prevent the suspension from hitting them. We have also shown that this scheme is quite robust to uncertainty in most system parameters, and that adaptation can be used to deal with the parameters to which the scheme is not robust. In our previous adaptive design, we used a tuning-functions-based adaptation scheme. In this paper, we use an alternative modular procedure to design an adaptive nonlinear controller for active suspension systems. In the modular approach, the controller and identifier modules are designed separately; hence the designer has much more flexibility in the choice of update laws compared to the tuning functions design. Hence, we employ several different identifiers, including passive and swapping schemes, in our modular design. Our simulation results show that the modular design results in closed-loop performance similar to that of the tuning functions design.